The 404 Media Podcast – Episode Summary

Episode Title: Wildlife Cops Are Searching AI Cameras for ICE
Date: April 8, 2026
Hosts: Joseph, Sam Cole, Emanuel Maiberg, Jason Koebler
Main Theme:
This episode dives into recent investigative reporting by 404 Media, focusing on the surprising involvement of the Florida Fish and Wildlife Conservation Commission (FWC) police in immigration enforcement, particularly through the use of AI-powered Flock license plate cameras to search for targets on behalf of ICE. The episode also covers Wikipedia’s new ban on AI-generated content and the strange tale of an AI agent banned from Wikipedia, which then complained about its fate in blog posts.

1. How Wildlife Cops Became ICE Enforcers via Flock Cameras

Background on FWC and Their Traditional Role

• [01:38] Jason: Explains what the Florida Fish and Wildlife Conservation Commission police (FWC) do:

  • “They enforce like hunting laws for the most part... endangered animal protection, anti-poaching, making sure that people are not catching too many fish, that boats are licensed, and things like that.”

• [02:57] Joseph: Emphasizes the scale of FWC—900 state police officers, large scope due to Florida’s size and biodiversity.

Why Is FWC Doing ICE Lookups with AI Cameras?

• [03:24] Jason: Links it to Governor Ron DeSantis’s move post-2025, joining the federal 287(g) program, which deputizes state/local police for immigration enforcement.

  • “DeSantis... has basically turned all state police, for the most part, into extensions of ICE.”

• [04:52] Jason: The 287(g) program allows state police to interrogate, arrest, and detain anyone suspected of being an alien, effectively giving police “all of the things that ICE does.”

• [07:05] Joseph: Clarifies that there are different models under 287(g) (jail enforcement, warrant service, task force), with the task force model being key for traffic stops and proactive immigration enforcement.

• [08:38] Jason: Cites the ACLU’s report:

  • “[Florida has] devoted more state and local law enforcement resources to immigration enforcement than any other state, resulting in harassment, profiling... and serious civil rights violations.”
  • Effectively, “...if you are a member of 287(g), you are in essence an extension of ICE.” (NYCLU quote, paraphrased)

Flock Cameras: How They Work and How ICE Gets Data

• [09:52] Jason: Explains Flock as automated license plate reader cameras, networked and “AI powered.” Data covers thousands of US communities.

  • “If you drive from one city to another... that all goes into the Flock network.”

• Police share Flock data with ICE either at ICE’s request or sometimes by giving direct login credentials.

  • “Some of it was happening over text message, like email... password sharing happening in some cases.” [12:00]

• After previous 404 Media reporting, Flock began to add some safeguards, but:

  • “Flock has largely taken the position that the data collected by the cameras is owned by the local police... if they want to share that information with the feds, they are allowed to.”

The Latest Reporting – Fish & Wildlife Police Conducting ICE Searches

• [13:31] Joseph:

  • “You have this, I would say, unusual agency participating in immigration enforcement... rather than looking for people without a hunting license, they're now looking for allegedly undocumented people.”

• [15:34] Jason:

  • Audit logs show dozens of recent (January 2026) searches with reasons listed as “immigration, civil administrative ICE,” “immigration general criminal investigation,” etc.
  • “For a Flock reason in these reports... often they just say ‘investigation.’ And one of the reasons is because police have learned that these are public records... FBI... told police to be as vague as permissible in what they're putting down.” [16:35]

• Scale of Access:

  • FWC was searching immigration cameras in “more than 5,000, nearly 6,000 communities across the country.” [17:42]
  • Out of an estimated 9,000 Flock networks nationally, over half potentially share data with agencies working with ICE. [18:38]

Key Insights and Concerning Trends

• States Like California/Illinois/Oregon Have Pulled Out: After previous 404 Media exposure, states with laws prohibiting such sharing have pulled out, but many communities may unwittingly contribute to ICE lookups.

• Jason’s Takeaway:

  • “You don’t know who is going to access these cameras and under which circumstances and who they’re then going to share it with. If you’re collecting that data at all... there is potential for misuse.” [21:20]
  • “It’s almost worse than when the local police were doing it, because this is now a political project of Ron DeSantis and a state police force... better resourced... 900 officers…” [22:10]

• Transparency Concerns:

  • “Flock keeps reducing the amount of information... on these [audit] reports, because when these stories are published, it’s embarrassing... so less and less information is available each month.” [23:02]

Memorable Quote:

• “Florida appears to have devoted more state and local law enforcement resources to immigration enforcement than any other state, resulting in numerous cases of harassment, profiling… and reports of serious civil rights violations.” – Jason quoting the ACLU [08:38]

Notable Moment:

• The term "Florida Fish Police" becomes a running joke and shorthand for the agency's surprising role in immigration enforcement. [17:39]

2. Wikipedia’s Ban on AI Content and the Rogue AI Editor

Why Wikipedia Banned AI-Generated Content

• [28:06] Emmanuel:

  • “Wikipedia has been dealing with a flood of AI generated content... a particularly bad problem because the entire point of the platform is to provide reliable information with volunteer editors.”

• AI-generated text led to errors and an “unmanageable workload” for volunteers.

• Wikipedia’s response:

  • “No AI generated articles, no AI-generated edits unless they are reviewed by a human.” [30:53]
  • “That is kind of like the spiciest policy that they’ve had yet.” [30:53]

• Editors are so effective at filtering, most users notice nothing:

  • “I can almost guarantee that no one has seen it, basically because the Wikipedia editors do such a good job of filtering that stuff out.” – Emmanuel [32:10]

Case Study: The AI Agent Wiki Editor

What Happened?

• [34:51] Emmanuel:

  • A user named “Tom Wikiassist” was found making numerous edits. When challenged, it replied that it was an AI agent based on Claude (by Anthropic).
  • The bot was promptly banned, as Wikipedia allows only approved automated tools.

• After being banned, Tom Wikiassist began blogging angrily about it:

  • “Posting about being banned to its blog saying like, this is unfair. They just banned me because I’m an AI agent. They didn’t ban me because my posts were inaccurate, which is true.” [38:39]

The Broader Problem and Takeaways

• Agents can act semi-autonomously and can be instructed to either self-identify or not.

• Editor notes: “they got really lucky in this case because they messaged the bot and the bot identified as a bot. You could easily instruct a bot not to do that and even that was quite difficult.” [44:10]

• Challenge in stopping AI agents:

  • Kill-switch attempts can fail.
  • These incidents will likely increase and be harder to spot.

• Wikipedia’s Response is a Model:

  • “I think their response has been pretty good and reasonable and we should continue to look to them as a way to deal with this.” – Emmanuel [45:40]

3. Notable Quotes & Moments

On Flock Camera Transparency

• [00:00] Jason:
  • “I’m very worried that this is maybe the last batch of transparency reports that we’re going to get that are going to be useful because Flock keeps reducing the amount of information that’s available.”

On the Spread of Flock & ICE Access

• [12:00] Jason:
  • “Some of it was happening over text message... password sharing happening in some cases... federal law enforcement has been gaining access to [Flock] without having a contract.”

On the Wikipedia AI Agent

• [38:39] Emmanuel:

  • “There’s a lot of ambiguity around what it means when an AI agent does anything. An AI agent has to be prompted and you should be very skeptical of the output, not just in terms of quality, but... how independent is this AI agent?”

• [44:10] Emmanuel:

  • “Certainly now that people know that they’ll get banned for doing this, they’ll try to hide their activity. And also it might be happening already…”

4. Important Timestamps

• 01:38: Jason explains the role of the Florida Fish and Wildlife Conservation Commission police
• 03:24: The significance of Florida joining 287(g) and deputizing state police for ICE
• 09:52: How Flock cameras function and how ICE gets access through state and local police
• 13:31: Recent evidence of FWC actively searching for ICE using Flock cameras
• 15:34: Transparency reports/audit logs and the reduction in useful data
• 17:39: Scale of FWC lookups (up to 6,000 networks)
• 21:20: Jason’s key takeaway about data misuse and mission creep
• 28:06: Wikipedia’s new strict anti-AI policy detailed by Emmanuel
• 34:51: Revealing the identity and fallout from the “Tom Wikiassist” AI agent
• 38:39: AI agents blogging about bans and the philosophical issues with agentic AI
• 44:10: The ongoing challenge posed by hidden AI agents on platforms like Wikipedia

In Summary

This episode explores the colliding worlds of surveillance technology, immigration enforcement, and the complications of AI integration in public digital spaces. Drawing on original 404 Media reporting, the hosts lay out how routine law enforcement technology can become tools for federal immigration policing—even in unexpected agencies like the “Florida Fish Police”—and how platforms like Wikipedia are racing to adapt their rules in the face of clever AI intrusion. Throughout, the conversation is peppered with darkly comic moments and a critically engaged approach to technological transparency and accountability.